In publications and presentations that use results obtained on this system, please include the following acknowledgement: “This work utilizes resources supported by the National Science Foundation’s Major Research Instrumentation program, grant #1725729, as well as the University of Illinois at Urbana-Champaign”.

System Description

User Guide

To request an account: email kindrtnk@illinois.edu

To access the system, submit jobs, etc.: Access hal-dgx and overdrive with hal-login3 Node

To compile CUDA applications, first enable devtoolset 9

scl enable devtoolset-9 bash

To work with Python, create and enable virtual environment first

python3 -m venv python3-virtualenv
source python3-virtualenv/bin/activate

You can use pip to install python packages within this environment.

To start a Jupyter notebook on hal-dgx

• on hal-dgx:

source python3-virtualenv/bin/activate
pip install jupyter # this needs to be done only once
jupyter notebook  # this will start the jupyter server

• on your own computer:

# This opens a connection to the hal-dgx.ncsa.illinois.edu Jupyter server, and 
# forwards any connection to port 8888 on the local machine to port 8888 on hal-dgx.ncsa.illinois.edu.
ssh -L 8888:localhost:8888 <userid>@hal-dgx.ncsa.illinois.edu

• Finally, on your own computer, open web browser and point it to the address you see after running 'jupyter notebook' on hal-dgx, something like http://localhost:8888/?token=...

• First, make sure you have a running Slurm job on hal-dgx

• on your own computer:

  ssh -L 8888:hal-dgx.ncsa.illinois.edu:21345 hal-login3.ncsa.illinois.edu

  Where 8888 is a port on your local computer, and 21345 is a port on hal-dgx you want to use for Jupyter

• on hal-dgx, start Jupyter using the port in the previous step
  • If the port is taken, repeat and try a different port